The desired impact of red blood cell (RBC) transfusion is to increase oxygen delivery as a means to protect tissues from ischemia. However, controversy exists over the safety and efficacy of transfusion with RBCs banked for short vs. long storage intervals. Although some retrospective studies in cardiac surgery patients have demonstrated associations between increased duration of RBC storage and increased mortality and morbidity, others have not detected any differences. No large, randomized prospective human study has attempted to correlate age of transfused RBCs with measures of end-organ dysfunction or death. PARENT STUDY: The REd CEll Storage Duration Study (RECESS) is being conducted by the NIH funded Transfusion Medicine and Hemostasis Clinical Trials Network. The primary objective is to determine if there are clinically important differences in outcomes and mortality in cardiac surgical patients (n=1,612) randomized to receive RBCs stored for shorter (d 10 days) vs. longer (e 21 days) periods of time. The impact of RBC storage on physiologic variables related to tissue oxygenation and the microcirculation has not been rigorously studied, nor have they been correlated with clinical outcomes. A few studies have investigated the effect of RBC transfusion on tissue oxygenation and the microcirculation. Of most relevance is a small study that demonstrated a deterioration in tissue oxygenation in trauma patients randomized to receive long storage RBCs but no change in those receiving short storage RBCs. PROPOSED ANCILLARY STUDY: Financial constraints limited us from pursuing the impact of storage duration on tissue oxygenation and the microcirculation in the parent trial. Our ancillary study proposes to obtain these additional measurements (e.g. peripheral and cerebral tissue oxygenation, and microvascular perfusion) at 4 study sites from a minimum of 250 participants in RECESS. Specific Aim: For optimal tissue viability, there must be adequate perfusion of the microcirculation and off-loading of oxygen by erythrocytes. We hypothesize that transfusion of RBCs stored for a longer time period decreases tissue oxygenation and microcirculatory perfusion compared with transfusion of RBCs stored for a shorter time period. This ancillary study will test this hypothesis by measuring tissue oxygenation (peripheral and cerebral) and microcirculatory perfusion following the transfusion of RBC that have been stored for shorter or longer periods of time in cardiac surgery patients enrolled in the RECESS trial. Raw electronic data will be obtained at 4 time points: prior to surgical incision, immediately prior to and after the unit of RBC transfusion in the ICU, and at 24 hours postoperatively. We will compare arms with respect to acute changes in these endpoints (change immediately prior to and after the administration of one unit of RBCs in the ICU) as well as less acute changes resulting from cumulative RBC units (change from prior to skin incision to 24 hours postoperatively). Raw data will be sent to an experienced core laboratory at Harvard Medical School, Boston, MA and will be analyzed by individuals blinded to study group assignment.